Optimized optics and feedback-controlled microscope hardware permit efficient acquisition of large, high- quality image datasets. Computer-based analyses deliver fast processing of high volume of image data which manually cannot be accomplished. The most exciting contribution that computer vision systems can make to translational cancer research, however, is to give access to image-based information that is inaccessible by eye. Computer vision programs can be directly coupled to mathematical models that describe the relation between hidden, invisible processes and measurable image events. Changes in the behavior of hidden processes are thus detectable as changes in the image. This study envisages the use of such algorithms to obtain statistically representative results for the differential effects of each of the three FDA-approved taxanes on the microtubule cytoskeleton in prostate cancer (PC) cell lines. My previous work in basic research has demonstrated the ability of computer-based analysis of the microtubule (MT) cytoskeleton to distinguish between weak disease phenotypes and establish links to MT dynamics in renal cell carcinoma. Therefore, the proposed translational research project can impact clinical decision-making by equipping physicians for the first time with a computer-aided tool allowing the design of an effective personalized MT-targeting chemotherapy of metastatic PC patients. Metastatic PC is treated primarily by means of taxane-based chemotherapy with one of the three FDA- approved taxanes (paclitaxel, docetaxel and cabazitaxel). However, currently there is no way of selecting the taxane for chemotherapy based on the particular pattern of dynamic behavior of the MT cytoskeleton in individual patients. In addition, recent data have indicated that AR binds MTs in order to traffic to the nucleus and that there are several clinically relevant AR splice variants i metastatic PC patients. To date, there is no information available on the potential effects of wild type or variant AR on MT dynamics and consequently no information on differential metastatic PC cell response to taxane treatment as a function of cellular AR content. Based on preliminary research, we hypothesize that there are inherent differences in tumor MT dynamics among individual PC patients, and that the presence of AR variants affects specific parameters of MT polymerization dynamics. If correct, this hypothesis has very significant implications for PC treatment. Because different microtubule-targeting drugs (even from within the same class like the taxanes) affect distinct parameters of MT dynamics, it is conceivable that we can match each drug with an individual tumor-specific MT-dynamics signature for maximum therapeutic efficacy.